* WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
@ 2009-02-10 12:51 Tomasz Chmielewski
2009-02-10 13:20 ` Dmitri Monakhov
2009-02-10 14:20 ` Jeff Mahoney
0 siblings, 2 replies; 4+ messages in thread
From: Tomasz Chmielewski @ 2009-02-10 12:51 UTC (permalink / raw)
To: LKML
I've seen this warning on 2.6.29-rc3 when I removed a iSCSI device without
unmounting a filesystem on it:
Aborting journal on device sdb.
------------[ cut here ]------------
WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
Hardware name: SCENIC EDITION
Modules linked in: ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext3 jbd tun kvm crc32c af_packet 8021q garp stp ipv6 binfmt_misc loop dm_mod cpufreq_ondemand cpufreq_conservative cpufreq_powersave p4_clockmod speedstep_lib usbhid freq_table fuse usb_storage snd_intel8x0 snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_pcm ppdev snd_timer snd_mixer_oss snd tulip soundcore snd_page_alloc ehci_hcd shpchp sis_agp ohci_hcd pci_hotplug sr_mod agpgart sg parport_pc parport rtc_cmos container thermal button processor evdev usbcore ide_generic ata_generic ide_pci_generic pata_acpi sis5513 ide_gd_mod ide_core pata_sis libata sd_mod scsi_mod crc_t10dif ext4 jbd2 crc16 [last unl
oaded: scsi_transport_iscsi]
Pid: 2726, comm: kjournald Not tainted 2.6.29-0.rc3.1mdv #1
Call Trace:
[<c038a02d>] ? printk+0x18/0x1b
[<c0133213>] warn_slowpath+0x73/0xd0
[<c0177953>] ? find_get_page+0x23/0xb0
[<c0133a69>] ? release_console_sem+0x1c9/0x200
[<c01c619c>] mark_buffer_dirty+0x7c/0x90
[<f85e746d>] journal_update_superblock+0x6d/0xd0 [jbd]
[<f85e7626>] __journal_abort_soft+0x86/0xa0 [jbd]
[<f85e6e6e>] ? __journal_remove_journal_head+0x9e/0x120 [jbd]
[<f85e7648>] journal_abort+0x8/0x10 [jbd]
[<f85e4dfd>] journal_commit_transaction+0xd8d/0xf50 [jbd]
[<c038a5de>] ? schedule+0x3de/0x9d0
[<c013d91a>] ? try_to_del_timer_sync+0x4a/0x60
[<f85e7e7d>] kjournald+0xad/0x1e0 [jbd]
[<c0148160>] ? autoremove_wake_function+0x0/0x50
[<f85e7dd0>] ? kjournald+0x0/0x1e0 [jbd]
[<c0147dec>] kthread+0x3c/0x70
[<c0147db0>] ? kthread+0x0/0x70
[<c0104867>] kernel_thread_helper+0x7/0x10
---[ end trace 66edd8e45842a174 ]---
journal commit I/O error
sd 6:0:0:1: [sdb] Synchronizing SCSI cache
--
Tomasz Chmielewski
http://wpkg.org
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
2009-02-10 12:51 WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90() Tomasz Chmielewski
@ 2009-02-10 13:20 ` Dmitri Monakhov
2009-02-10 14:11 ` Tomasz Chmielewski
2009-02-10 14:20 ` Jeff Mahoney
1 sibling, 1 reply; 4+ messages in thread
From: Dmitri Monakhov @ 2009-02-10 13:20 UTC (permalink / raw)
To: Tomasz Chmielewski; +Cc: LKML
Tomasz Chmielewski <mangoo@wpkg.org> writes:
> I've seen this warning on 2.6.29-rc3 when I removed a iSCSI device without
> unmounting a filesystem on it:
Block device has just burned under filesystem's feet (blkdev return
EIO for all requests.) What are you expect in this condition?
As soon as i understand your systems is alive without any BUG or OOPS.
The only consequence is big-bada-boom message which is IMHO quite right.
>
> Aborting journal on device sdb.
> ------------[ cut here ]------------
> WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
> Hardware name: SCENIC EDITION
> Modules linked in: ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core
> ib_addr iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext3 jbd
> tun kvm crc32c af_packet 8021q garp stp ipv6 binfmt_misc loop dm_mod
> cpufreq_ondemand cpufreq_conservative cpufreq_powersave p4_clockmod
> speedstep_lib usbhid freq_table fuse usb_storage snd_intel8x0
> snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event
> snd_seq snd_seq_device snd_pcm_oss snd_pcm ppdev snd_timer
> snd_mixer_oss snd tulip soundcore snd_page_alloc ehci_hcd shpchp
> sis_agp ohci_hcd pci_hotplug sr_mod agpgart sg parport_pc parport
> rtc_cmos container thermal button processor evdev usbcore ide_generic
> ata_generic ide_pci_generic pata_acpi sis5513 ide_gd_mod ide_core
> pata_sis libata sd_mod scsi_mod crc_t10dif ext4 jbd2 crc16 [last
> unloaded: scsi_transport_iscsi]
> Pid: 2726, comm: kjournald Not tainted 2.6.29-0.rc3.1mdv #1
> Call Trace:
> [<c038a02d>] ? printk+0x18/0x1b
> [<c0133213>] warn_slowpath+0x73/0xd0
> [<c0177953>] ? find_get_page+0x23/0xb0
> [<c0133a69>] ? release_console_sem+0x1c9/0x200
> [<c01c619c>] mark_buffer_dirty+0x7c/0x90
> [<f85e746d>] journal_update_superblock+0x6d/0xd0 [jbd]
> [<f85e7626>] __journal_abort_soft+0x86/0xa0 [jbd]
> [<f85e6e6e>] ? __journal_remove_journal_head+0x9e/0x120 [jbd]
> [<f85e7648>] journal_abort+0x8/0x10 [jbd]
> [<f85e4dfd>] journal_commit_transaction+0xd8d/0xf50 [jbd]
> [<c038a5de>] ? schedule+0x3de/0x9d0
> [<c013d91a>] ? try_to_del_timer_sync+0x4a/0x60
> [<f85e7e7d>] kjournald+0xad/0x1e0 [jbd]
> [<c0148160>] ? autoremove_wake_function+0x0/0x50
> [<f85e7dd0>] ? kjournald+0x0/0x1e0 [jbd]
> [<c0147dec>] kthread+0x3c/0x70
> [<c0147db0>] ? kthread+0x0/0x70
> [<c0104867>] kernel_thread_helper+0x7/0x10
> ---[ end trace 66edd8e45842a174 ]---
> journal commit I/O error
> sd 6:0:0:1: [sdb] Synchronizing SCSI cache
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
2009-02-10 13:20 ` Dmitri Monakhov
@ 2009-02-10 14:11 ` Tomasz Chmielewski
0 siblings, 0 replies; 4+ messages in thread
From: Tomasz Chmielewski @ 2009-02-10 14:11 UTC (permalink / raw)
To: Dmitri Monakhov; +Cc: LKML
Dmitri Monakhov schrieb:
> Tomasz Chmielewski <mangoo@wpkg.org> writes:
>
>> I've seen this warning on 2.6.29-rc3 when I removed a iSCSI device without
>> unmounting a filesystem on it:
> Block device has just burned under filesystem's feet (blkdev return
> EIO for all requests.) What are you expect in this condition?
Ask someone not familiar with kernel internals which message tells more:
WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
or:
WARNING: Block device has just burned under filesystem's feet
> As soon as i understand your systems is alive without any BUG or OOPS.
> The only consequence is big-bada-boom message which is IMHO quite right.
The "big-bada-boom message", "---[ cut here ]---" don't explain what
happened, but rather "indicate a major problem" (as it's printed by
kernel/panic.c).
--
Tomasz Chmielewski
http://wpkg.org
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90()
2009-02-10 12:51 WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90() Tomasz Chmielewski
2009-02-10 13:20 ` Dmitri Monakhov
@ 2009-02-10 14:20 ` Jeff Mahoney
1 sibling, 0 replies; 4+ messages in thread
From: Jeff Mahoney @ 2009-02-10 14:20 UTC (permalink / raw)
To: Tomasz Chmielewski; +Cc: LKML
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Tomasz Chmielewski wrote:
> I've seen this warning on 2.6.29-rc3 when I removed a iSCSI device without
> unmounting a filesystem on it:
You can ignore this warning. There's a long-standing issue where it will
print warnings after the superblock has failed to be written out. When
that happens, the regular endio handler marks it !uptodate. The warning
occurs when another attempt is made to write it out since it's not
marked uptodate again. The superblock is special in that it is read once
at mount, and then write-only afterwards, so the concept of !uptodate
doesn't really apply.
I posted a patch to work around this some time ago, but it's only a
temporary fix. I suspect that nobody's tackled this because the handling
is in jbd and the superblock is a special case.
- -Jeff
> Aborting journal on device
> sdb.
> ------------[ cut here
> ]------------
> WARNING: at fs/buffer.c:1250
> mark_buffer_dirty+0x7c/0x90()
> Hardware name: SCENIC
> EDITION
> Modules linked in: ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core
> ib_addr iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext3 jbd
> tun kvm crc32c af_packet 8021q garp stp ipv6 binfmt_misc loop dm_mod
> cpufreq_ondemand cpufreq_conservative cpufreq_powersave p4_clockmod
> speedstep_lib usbhid freq_table fuse usb_storage snd_intel8x0
> snd_ac97_codec ac97_bus snd_seq_dummy snd_seq_oss snd_seq_midi_event
> snd_seq snd_seq_device snd_pcm_oss snd_pcm ppdev snd_timer snd_mixer_oss
> snd tulip soundcore snd_page_alloc ehci_hcd shpchp sis_agp ohci_hcd
> pci_hotplug sr_mod agpgart sg parport_pc parport rtc_cmos container
> thermal button processor evdev usbcore ide_generic ata_generic
> ide_pci_generic pata_acpi sis5513 ide_gd_mod ide_core pata_sis libata
> sd_mod scsi_mod crc_t10dif ext4 jbd2 crc16 [last unloaded:
> scsi_transport_iscsi]
> Pid: 2726, comm: kjournald Not tainted 2.6.29-0.rc3.1mdv
> #1
> Call
> Trace:
> [<c038a02d>] ?
> printk+0x18/0x1b
> [<c0133213>]
> warn_slowpath+0x73/0xd0
> [<c0177953>] ?
> find_get_page+0x23/0xb0
> [<c0133a69>] ?
> release_console_sem+0x1c9/0x200
> [<c01c619c>]
> mark_buffer_dirty+0x7c/0x90
> [<f85e746d>] journal_update_superblock+0x6d/0xd0
> [jbd]
> [<f85e7626>] __journal_abort_soft+0x86/0xa0
> [jbd]
> [<f85e6e6e>] ? __journal_remove_journal_head+0x9e/0x120
> [jbd]
> [<f85e7648>] journal_abort+0x8/0x10
> [jbd]
> [<f85e4dfd>] journal_commit_transaction+0xd8d/0xf50
> [jbd]
> [<c038a5de>] ?
> schedule+0x3de/0x9d0
> [<c013d91a>] ?
> try_to_del_timer_sync+0x4a/0x60
> [<f85e7e7d>] kjournald+0xad/0x1e0
> [jbd]
> [<c0148160>] ?
> autoremove_wake_function+0x0/0x50
> [<f85e7dd0>] ? kjournald+0x0/0x1e0
> [jbd]
> [<c0147dec>]
> kthread+0x3c/0x70
> [<c0147db0>] ?
> kthread+0x0/0x70
> [<c0104867>]
> kernel_thread_helper+0x7/0x10
> ---[ end trace 66edd8e45842a174
> ]---
> journal commit I/O
> error
> sd 6:0:0:1: [sdb] Synchronizing SCSI cache
>
>
- --
Jeff Mahoney
SUSE Labs
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org
iEYEARECAAYFAkmRjSMACgkQLPWxlyuTD7KXtwCgl5zQ569WVOr6aqp+l/P5kGl4
cbgAni+SmrTeTz0jY5lU763TlfncqxdU
=uY+Z
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2009-02-10 14:22 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-02-10 12:51 WARNING: at fs/buffer.c:1250 mark_buffer_dirty+0x7c/0x90() Tomasz Chmielewski
2009-02-10 13:20 ` Dmitri Monakhov
2009-02-10 14:11 ` Tomasz Chmielewski
2009-02-10 14:20 ` Jeff Mahoney
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.